Dead is dead. Who care about "gun" deaths? The question should be whether or not increased firearms ownership causes more deaths in general.

After having done a fair bit of academic reading on the subject, my overall impression of the research is that gun ownership is not correlated with homicide to a statistically significant degree, positive or negative. Other factors strongly correlate with homicide rate, like income inequality, and entirely explain why the US has crime rates closer to countries like Brazil than to countries like Finland. However, gun ownership does have a significant positive correlation to suicide. I'll leave the policy implications to another time, except to say that it's clear that in our history with gun control policy we have consistently attempted to address the wrong problem.

A) The action of WP is much different than that of napalm, even though both are components of incendiary weapons. Both are designed to ignite fires, but napalm is intended to "stick" and provide a persistent fuel source over a wide area. Regardless, in modern practice WP is used primarily for smokescreening and target marking.

B) In any case, neither napalm nor WP are considered "chemical weapons" by any treaties--there are treaties that regular incendiary weapons (e.g. Convention on Certain Conventional Weapons and the Geneva Conventions), and these may or may not cover WP weapons, but WP is not used for toxic effect (like sarin or VX) and is not covered by the same restrictions (e.g. the Chemical Weapons Convention). Consequently, no state considers WP to be WMD but most would consider sarin as such. This seems reasonable given the effects and use of these weapons. All modern weapons contain "chemicals," like TNT, but that doesn't make them "chemical weapons."

Orbital Sciences is running the mission--for whatever reason, they've set up shop just outside Wallops and have been spending quite a bit of time and money getting one of the pads at Wallops set up for Minotaur and Antares launches. It could be because it's closer to their manufacturing facilities, because they've been launch at WFF with smaller rockets many years, because the costs of using the pad and facilities at WFF are cheaper than the big pads at KSC, or whatever; but the point is that Orbital is running the mission and almost certainly chose to use WFF themselves. While it's true they might get better performance launching from KSC, it would shock me if WFF wasn't much cheaper to run out of. But even if it weren't, Orbital has invested in their operations at Wallops and is very unlikely move everything down to Florida now. I don't know what the deal is here, but in this case I suspect it's just a business decision by Orbital and nothing more sinister.

It's interesting that Fortran allows and can support GOTO from a variable line number; do you have a *good* example of why one might want to do this?

You can add binary repos to Macports, if you wish. Macports is trying to provide something like BSD ports on OSX, though, so compiling from source is the default behavior. I will say that for my purposes, Macports has been less hassle than either Fink or Brew. It has the the best variety in software and works fairly reliably. All three, unfortunately, suffer from the fact that they don't control the system so OS and Xcode updates can and do break packages.

This is simply false. Selective Availability is turned off, meaning the accuracy is no longer intentionally degraded on the unencrypted L1 channel, but the encrypted L2 signal is still broadcast and it is still restricted to military equipment precisely to prevent spoofing. GPS Block III will add an encrypted L2C signal so that civilian airliners and other safety-critical applications can have access to a spoof-resistant GPS signal.

Seems to me a lot of what you want doesn't fit into the vision of KA. You want to be able to control what content the students see and when they see it, so that you can keep the entire class working on the same stuff at the same time and at the same pace. That's in opposition to what I understand to be Khan's goal of encouraging self-directed learning, where students learn at their own pace, ability, and interest. So, the long and short of it is that I wouldn't hold my breath waiting for KA to implement these features given their limited resources and goals.

This is true, but from my experience anything involving "signatures" is much more difficult to share with foreigners, including our closest allies. I've had trouble sharing signature data between allies operating exactly the same equipment, let alone something one party has but the other does not. I doubt F-22 emissions data is shared at all, at any level.

I always thought Opera's selling points had more to do with the customizable UI, and not so much to do with it's rendering engine. For your typical end-user, the impact of Opera having a unique rendering engine is that some pages look funny in Opera because few websites test against Presto. Webkit can't be ignored these days, so by adopting it Opera has less to worry about and gets an engine that keeps up-to-date at a much lower cost than rolling their own. Moreover, should Opera wish to add features to the rendering engine (such as new proposals for the HTML spec) it's a hell of a lot easier to get other to adopt them when you've implemented it in Webkit than a closed source implementation in Presto. It's a win for everyone; the only surprise to me is that they didn't adopt an open renderer sooner.

This is exactly right. I myself earned a PhD in Aerospace Engineering at a top research university--I'm fortunate to have an American-born mother and have been able to continue to work on my research in the US after earning my degree, but more of my classmates were foreign students. Not only do they have a difficult time getting green cards after graduating, but they're unable to find jobs in the US because most Aerospace jobs are defense or government related and the US government has made it very difficult for foreign nationals to work in US research centers or for defense contractors. So after being trained to do research in militarily useful areas, and after they try and fail to find jobs in the US, these students are forced to go back home and end up contributing to foreign defense industries.

Now, you might say that US graduate programs should hire on more US graduate students--but the reality is, there just aren't that many qualified students applying to engineering graduate programs and bringing in students from abroad massively increases the selection pool and therefore quality of grad students available to faculty. That students the world over want to, and are allowed to, enter US graduate programs is a big part of why US academic research is still the envy of the world, even though the US has declined from the top spot in many other areas. Grad students, who are really just "research apprentices" do most of the research work coming out of universities, with faculty supervising, directing, and selling research programs. Limiting foreign admission would be an enormous blow to US STEM academia and our overall research output. The best option is to figure out how we can retain the best foreign students in the US. One solution might be for the US government to expand it's national labs greatly, admit foreign nationals on the path the citizenship, and directly support basic research outside of the academic system. More US jobs in basic research might also stimulate more US students to apply to graduate programs in STEM, and overall, would help the US to maintain it's gradually decaying edge in the high tech sector.

I've tried several times to migrate to Octave, but it's still just not good enough of a "MATLAB-clone" to really replace MATLAB for my purposes. I run MATLAB through EMACS matlab-mode, so I don't care about the editor or GUI.

What keeps me away from Octave is:
1) MATLAB is much faster at solving most problems.
2) Much of the capability available in Mathworks supported toolboxes is missing from Octave.
3) Handle graphics capabilities are now available in Octave, but lag far behind MATLAB. I don't write MATLAB GUIs, but many others do, and I sometimes need to run their code.

On the other hand, scientific computing support in Python is pretty good and rapidly improving. If I just wanted capability "like" MATLAB, I'd use Python. However, in most engineering R&D environments, MATLAB support is important for collaboration, and here is where Octave could be useful were it more comparable to MATLAB.

Fine, but by a similar arguments space exploration as a whole, manned or unmanned, is basically worthless. Learning about Martian soil is unlikely to, in and of itself, provide any economic or "practical" payback to those of us here on Earth in the utilitarian sense you seem to advocate. Instead of worrying about space, we should allocate our money to things with more immediate and predicable impact; there's no shortage of opportunities in medical research or technological development in energy, transportation, or computing where money could be better spent by your measure.

Yet, clearly adventure, wonder, art, and exploration are highly valued by most people--a huge portion of our privately funded economic activity is directed towards these things which have seemingly no economic value, and has been throughout recorded history. And just like more "practical" areas, sometimes there are projects so big and so risky that no single private entity can take them on, but that doesn't mean they're wasteful. You don't here anyone say that they sure wished the Byzantine's didn't waste their resources building the Hagia Sophia or the Americans the Apollo project, because while there's no obvious practical benefits to either which couldn't have been met more efficiently though direct spending towards practical applications, both had immeasurable cultural value.

I feel the same way about manned spaceflight today--the only good reason to explore the universe is for the adventure, because it's exciting, interesting, inspiring, and enlightening. In short, because it makes us better. And I see no better way to meet that objective than through manned spaceflight.

Expecting space science, whether conducted by robot or man, to cure the sick or increase economic productivity is like expecting to buy a winning lottery ticket. So if science is your only justification for space flight you shouldn't expect space flight to last much longer because we can do "better" science for cheaper right here on Earth, and there's no shortage of ideas to fund.

Quite correct--new technology development is a good thing for the military, the taxpayer, and of course US industry. But we need to arrest "Augustine's Law," the sort of Moore's law for the cost of US military aircraft. New aircraft should be designed to be much cheaper than the old aircraft they replace, and new technology should be used to make design trades towards economy, not just performance as in the past. As new aircraft come online, we need to abandon older vehicles with high maintenance costs. In the political realm, we also need to reorient our objectives for the military--in the long term the US will no longer be the overwhelmingly dominant player in the Western world (not to say it won't be important, even the most important). Accordingly, the US should not aim to be the overwhelmingly dominant military force for the West; let's get our NATO allies to build the capability to project power. For example, the next time France and the UK want to intervene in North Africa, they should be able to do it without the logistical support of the US.

You're wrong about the V-22; the development program was long delayed and over budget, and several complications arose during testing. However, since operational deployment, the aircraft has an excellent safety record--in fact much better than the CH-47s it's been replacing. Your opinion on the aircraft is dated, and doesn't represent how things have turned out in actual practice. And unlike the F-22, the V-22 is very well suited to recent types of conflicts--it can rescue downed pilots or deploy a special forces assault team more quickly and over a longer range the a conventional helicopter (at the cost of less payload and more complexity).

The F-22 is a different beast, but we've already stopped production and now have these airframes. We should work out the issues, and there's nothing fundamental to say we can't do so, and then take older airframes offline once the F-22 is fit for combat operations as newer airframes are cheaper to maintain than older ones, even the F-22. We should probably also build a minimal complement of F-35, and then actually replace the remainder of our combat aircraft resulting in a much smaller total fleet, sized more to our needs. The US is losing its status as hegemon of the Western world, and look to itself as just one of a bunch of Western democracies--part of this is to scale back militarily and let the other NATO countries take a bigger role in their own military defense and power projection.

Lastly, as for UAVs the days are over where they are necessarily cheaper than manned aircraft--so much capability is demanded from these aircraft that in the end, they're now costing as much as manned vehicles, and consequently corners can't be cut on reliability, either. Vehicles like the Predator and Global Hawk are actually requiring more ground crew than comparable manned aircraft in order to interpret incoming data and act on it effectively. They will play a bigger role in the future, for sure, but don't think building UCAVs to replace F-22s will save a single penny.

Steamworks DRM is an option, but not a requirement, for anyone who uses Steamworks when developing their game. And it's fairly unobtrusive. Some games will use both Steamworks and more invasive DRM. But you're right, many games come through Steam totally DRM free. For instance, you can just copy many of the Paradox Interactive titles from their install folder and put them on a different computer, and they'll just run.